X-ray powder diffraction (XRD) measurement
The XRD patterns of synthesized bare Fe3O4, Fe3O4–LCysteine, and Fe3O4–LCysteine–GQDs nanostructures in comparison with the standard pattern of Fe3O4 nanoparticles (JCPDS data, card no. 96-900-5815), and Fe2O3 nanoparticles (JCPDS data, card no. 96-101-1241) are presented in Fig. 2. According to the standard pattern of magnetite nanoparticles, which is shown in Fig. 2a, the well-match pattern can be observed in Fig. 2c and confirms the inverse spinel and face-centered cubic (FCC) structure of synthesized magnetite particles which is belonging to the Fe3O4 nanoparticles. The average crystalline size of the Fe3O4 nanoparticles was determined using the Debye–Scherrer formula and full-width at half-maximum (FWHM) of the three most intense peaks (Ahmadi and Hosseini 2013):
$$D = \frac{0.9 \lambda }{{\beta {\text{Cos}} \left( \theta \right)}},$$
(2)
where D is the mean crystalline size, l is the X-ray wavelength, b is the FWHM (radian) of diffraction peak, and \(\theta\) is diffraction angle. According to the formula and diffraction patterns, the average crystal size of the Fe3O4 particles was calculated by about 7 nm. Figure 2d, e confirms that the presence of LCysteine and GQDs did not considerably alter XRD patterns of the bare MNPs, so the nature and phase structure of the MNPs remains constant; however, it could be seen that some insignificant peaks with low intensity appear in the functionalized sample. The presence of GQDs can change some of the iron oxide nanoparticles’ structure from Fe3O4 to Fe2O3. As a result, another peak at 33.35°, which is related to the (104) appears and confirms the formation of a Fe2O3.
Fourier-transform infrared (FT-IR) spectroscopy
The FT-IR spectra and characterization bands of Fe3O4–LCysteine–GQDs nanocomposites compare to the naked Fe3O4 nanoparticles’ spectra, pure LCysteine, and Fe3O4–LCysteine are shown in Fig. 3. Figure 3a shows the FT-IR spectra of uncoated Fe3O4 nanoparticles. The broad peak at 3423.40 cm−1 attributes to the presence of stretching and bending vibrations of the O–H bond and this is due to the entrapped moisture and adsorbed water. The two peaks at 572.82 cm−1 and 632.60 cm−1 are assigned to iron oxide nanoparticles’ Fe–O bond vibration. The other peaks are related to the alkyl stretching modes, which were used during the synthesis procedure for pH rise. Figure 3b represents the \({\text{L}}_{{{\text{Cysteine}}}}\) FT-IR spectra. Typically, most of the amino acids can exist as zwitterions, exhibiting both carboxylic (-COOH) and amino (-NH2) functional groups (Sangeetha and Philip 2013). As mentioned above, LCysteine is a neutral amino acid, so the FT-IR spectrum of the commercial LCysteine shows broad characteristics range band of NH3+ stretching mode at 3000 cm−1, and the N–H bond of the amino group is observed at 1501.05 cm−1. The presence of the thiol group is virtually confirmed with the band at 2069.31 cm−1 (Hai et al. 2018a, b; Tiwari 2017). LCysteine coated Fe3O4 nanoparticles show characteristic bands in FT-IR spectroscopy in Fig. 3c. These spectra exhibit insignificant changes in characteristic bonds compared with pristine LCysteine, shown in Table 1. The absence of the S–H band at 2069.31 cm−1 strongly shows that magnetite nanoparticles have been linked with the thiol group of LCysteine to form a core–shell structure (Cohen 2008; Tiwari 2017; Karamipour 2015). The Fe–S peak cannot be seen in Fig. 3c, since Fe–S stretching, and bending peaks appear below 400 cm−1 (Lauterbach 2015). Finally, the FT-IR spectra of the final nanocomposite are depicted in Fig. 3d. A decrease in N–H characteristic band intensity can result in forming the bond between the amino acid functional group of LCysteine and GQDs (Hasanzadeh 2016; Mohammad-rezaei 2014). The wide peak at 3000 cm−1 for the samples, including LCysteine is related to the NH3+ stretching mode. The other characteristic bands with their wavenumber are shown in Table 1.
Table 1 The list of characteristic band wavenumber of LCysteine, LCysteine coated MNPs, and Fe3O4-LCysteine-GQDs X-ray photoelectron spectroscopy (XPS) analysis
XPS measurement leads to a deeper understanding of the nature of the bonding types between LCysteine and Fe3O4 nanoparticles. The results of the XPS analysis of LCysteine functionalized Fe3O4 is shown in Fig. 4a, d. There are 5 distinct, sharp peaks at 710 eV, 531 eV, 401 eV, 288 eV, and 163 eV in the full spectrum, which are described as the Fe2p, O1s, N1s, C1s, and S2p, respectively (Fig. 4a). The possibility of forming the bonding of amino and carboxylic groups of LCysteine to the Fe3O4 nanoparticles can be neglected, since the singlet spectrum of N1s and O1s in the LCysteine functionalized Fe3O4 is detected, which are correlated to the amino and carboxylic groups of the LCysteine molecules. In the C1s spectrum in Fig. 4b, three peaks, which are detected at 288.2 eV, 287.6 eV, and 285.1 eV, ascribe the presence of C–C, C–N/C–H/C–OH, and C = O, respectively and show the existence of LCysteine in the final product. These spectra also announce that the carboxyl group in LCysteine molecules did not react with the iron oxide nanoparticles. Therefore, according to the most similar research, the most probable binding type is the formation of the Fe–S bond (Ahmadi 2011; Sangeetha and Philip 2013; Cohen 2008). The S2p XPS spectra of MNPs functionalized with LCysteine is observed in Fig. 4c. Instead of two peaks at 163.7 eV and 164.8 eV for pure LCysteine, other characteristic peaks appear at 161.2 eV and 162.8 eV, Which confirm the chemical bond of Fe–S in the core–shell structure and assure us that the LCysteine interacts with Fe3O4 nanoparticles via its thiol group (Cohen 2008). The peaks at 709.5 eV and 722.9 eV reveal the Fe2p spectrum for the functionalized MNPs and associate to the MNPs chemical bonds (Fig. 4d). Besides, compared with the bare MNPs, a doublet peak is detected at 717.19 eV and 731.4 eV when the particles coated with LCysteine. The presence of 2 additional peaks at Fe2p XPS spectra of core–shell structure compare to the bare MNPs is due to the doublet nature of the excited ions, where the final state is available with two possible energies, formed from a set of degenerated states. These states of energies depend on the total angular momentum quantum number j = l + s, where l is the orbital angular momentum number and s is the spin angular momentum quantum number.
Scanning electron microscopy (SEM) characterization
SEM images of Fe3O4, Fe3O4–LCysteine, and Fe3O4–LCysteine–GQDs nanostructures, shown in Fig. 5a–c, illustrate the homogenized distribution and sponge-like morphology of all samples in which the mean size of the particles is about 15–20 nm. Although the excellent stability and high dispersion of nanoparticles of magnetite in an aqueous solution, they aggregate as soon as they dry. Therefore, in this micrograph, each cluster has a number of smaller particles of magnetic nanoparticles. Furthermore, EDX measurements were used to confirm the composition of the synthesized samples (Fig. 5d–f). The quantity of the Fe, O, C, N, and S contents are shown in Table 2. The calculated elements’ content quantities reveal the successful deposition of the MNPs–LCysteine–GQDs nanocomposites.
Table 2 Elements’ content quantities of \({\text{Fe}}_{{3}} {\text{O}}_{{4}}\), \({\text{L}}_{{{\text{cysteine}}}} { - }\)coated \(Fe_{3} O_{4}\) nanoparticles and \({\text{Fe}}_{{3}} {\text{O}}_{{4}}\)-\({\text{L}}_{{{\text{Cysteine}}}}\)-\({\text{ GQDs}}\) nanocomposite measured by EDX spectra Transmission electron microscopy (TEM) characterization
TEM images in Fig. 6a refer to the naked Fe3O4 in the absence of LCysteine, and Fig. 6b is related to the modified Fe3O4 with LCysteine. As shown in Fig. 6a, bare iron oxide particles have a high tendency to aggregate due to their strong magnetic and long-range Vander Waal’s forces. Figure 6b shows Fe3O4 particles functionalized with LCysteine and its dispersion behavior. The core size of the particles has been increased during the functionalization process. It is clear from size distribution histogram diagrams in Fig. 6c, d, the mean particle size for Fe3O4 nanoparticles, Fe3O4–LCysteine core–shell structure, and LCysteine layer thickness are about 7.5 \(\pm\) 3 nm, 11 ± 4 nm, and 3 ± 1 nm, respectively.
Vibrating sample magnetometer (VSM) diagrams
Magnetic hysteresis curves of the as-prepared Fe3O4, Fe3O4–LCysteine, and Fe3O4–LCysteine–GQDs structures at room temperature are depicted in Fig. 7a–c. The values of saturation magnetization (Ms), remnant magnetization (Mr), and magnetic coercivity (Hc) for all of the prepared samples show a superparamagnetic behavior and reveal their small size (Table 3). As it can be observed from data in Table 3, the superparamagnetic behavior of the bare Fe3O4 nanoparticles is confirmed from the high Ms and low Hc, which are 56.5 emu. g−1 and 0.28 Oe, respectively. Additionally, Ms decreases to 53.3 emu. g−1 when LCysteine is coated MNPs, while Hc merely increases and reaches to 0.38 Oe. In other words, due to the core–shell interface and shielding effects, the LCysteine layer decreases the magnetic properties compared with the bare magnetite nanoparticles. Finally, magnetic saturation for the Fe3O4–LCysteine–GQDs nanocomposite decreases to 28.99 emu. g−1, while the Hc reaches 0.09 Oe. Although the presence of GQDs decreases the Ms, the nanostructure still shows superparamagnetic properties. The reduction in Ms value has been referred to the disordering of the magnetic moment. The higher the GQDs shell thickness onto the magnetite nanoparticles leads to the lower magnetization saturation. Additionally, according to the XRD analysis, the change in the structure of Fe3O4–Fe2O3 is another reason to cause a decrease in the Ms value. As a result of the nanocomposite’s magnetic–nonmagnetic hybridization, these results are still sufficient for MRI contrast agents.
Table 3 Magnetic properties of \(Fe_{3} O_{4}\), \({\text{L}}_{{{\text{cysteine}}}}\)-coated \({\text{Fe}}_{{3}} {\text{O}}_{{4}}\) nanoparticles and \({\text{Fe}}_{{3}} {\text{O}}_{{4}}\)-\({\text{L}}_{{{\text{Cysteine}}}}\)-\({\text{ GQDs}}\) nanocomposite Optical study
The UV–Vis spectra of GQDs and MNPs covered with GQDs lead to a strong absorption peak at 276 nm and 270 nm with a tail extending to the visible region, respectively (Fig. 8a, b). The high UV–Vis adsorption is due to the presence of the GQDs, which leads to the surface sphere and surface cavity mode at the interface of MNPs and GQDs. Photoluminescence (PL) spectra of the GQDs, MNPs, and Fe3O4–LCysteine–GQDs nanocomposites are recorded at room temperature (Fig. 8c–e). A detailed PL study for GQDs was carried out using different excitation wavelengths. Apart from most of the carbon-based nanomaterials with highly dependent PL excitation wavelength, GQDs show a different behavior (Mohammad-rezaei 2014; Li 2010). As it can be seen in Fig. 8c, the emission wavelength of GQDs is nearly excitation-independent (Zhuo et al. 2012). By getting deeper in Fig. 8c, it can be observed that as the excitation wavelength varies from 235 to 400 nm, GQDs emit a stable, strong PL peak at about 460 nm (2.69 eV). This phenomenon happens when the size and sp2 clusters on the surface of the GQDs remain constant and uniform (Dong 2012). Besides, the PL spectra of the bare MNPs revealed a stable emission at 530 nm (2.34 eV) when it is excited at 235 nm (Fig. 8d). The distinct PL features and different emission wavelengths in the visible range were measured with photoexcitation at 235 nm for the final product, which presents an asymmetric emission curve (Fig. 8d). Figure 8d shows a broad emission peak at around the visible range with two unequivocal adsorption peaks at about 460 nm and 530 nm. It is clear that the former peak nearly stands at the same position compare with the pristine GQDs, while the second peak can be ascribed to the presence of magnetic core in the final nanocomposites. As a result, the PL spectra of the final product shows a stable emission in the visible range, which makes this nanocomposite a promising candidate for bioimaging applications.